End-to-end Keyword Spotting using Xception-1d
Computation and Language
2021-10-15 v1
Abstract
The field of conversational agents is growing fast and there is an increasing need for algorithms that enhance natural interaction. In this work we show how we achieved state of the art results in the Keyword Spotting field by adapting and tweaking the Xception algorithm, which achieved outstanding results in several computer vision tasks. We obtained about 96\% accuracy when classifying audio clips belonging to 35 different categories, beating human annotation at the most complex tasks proposed.
Keywords
Cite
@article{arxiv.2110.07498,
title = {End-to-end Keyword Spotting using Xception-1d},
author = {Iván Vallés-Pérez and Juan Gómez-Sanchis and Marcelino Martínez-Sober and Joan Vila-Francés and Antonio J. Serrano-López and Emilio Soria-Olivas},
journal= {arXiv preprint arXiv:2110.07498},
year = {2021}
}
Comments
In proceedings of ESANN 2021 conference. 5 pages + references